Lipopolysaccharides (LPSs) are unusual glycolipids that make up the outer surface of the outer membranes of gram-negative bacteria, including strains of Rhizobium. Because of the difficulties encountered in working with these substances, the exact covalent structure of the hydrophobic anchor of LPS (termed lipid A) was unknown prior to 1983. An understanding of the structure, biosynthesis and function of lipid A is very important, because lipid A is the active components of LPS responsible for many of the clinical complications of gram-negative sepsis. For this reason lipid A is also known as endotoxin. Progress with the elucidation of the biosynthesis and molecular biology of endotoxins resulted from the P.I's. discovery in 1983 of monosaccharide lipid A precursors that accumulate in phosphatidylglycerol-deficient mutants of Escherichia coli. It is now established that lipid A is essential for the growth of gram-negative bacteria and that inhibitors of lipid A synthesis could be novel antibiotics. In addition, certain precursors and analogs of lipid A are antagonists of the action of lipid A as an endotoxin and may be prototypes for new pharmacological agents useful in the treatment of endotoxin-induced shock. Most studies of the biosynthesis and pharmacological properties of endotoxins have been based ont he structure of lipid A from E. coli, Salmonella minnesota and Rhodobacter sphaeroides. The proposed work seeks to elucidate the unique features of lipid A biosynthesis in Rhizobium leguminosarum. The structure of R. leguminosarum lipid A is provocative, as it lacks both the 1 and 4' phosphates that are commonly found in lipid As and are crucial in E. coli lipid A biosynthesis. Lipid A of R. leguminosarum also lacks the glucosamine disaccharide of E. coli lipid A. Instead, it contains a unique 2-deoxy-2-aminogluconate moiety in place of the reducing end glucosamine. Lipid A of Rhizobium is also acylated with long fatty acids not found in E. coli. The structure of lipid A in R. leguminosarum suggests the existence of novel enzymes that may be useful for the preparation of endotoxin analogs. Despite its structure, R. leguminosarum lipid A stimulates synthesis of cytokines in certain immune cells, showing that it has some of the biological properties of E. coli lipid A. R. leguminosarum lipid A and its precursors require further study to evaluate their possible utility as endotoxin mimics and/or antagonist. Identification of the genes and enzymes that generate the unique lipid A of R. leguminosarum may also provide new insights into the function of lipid A-like molecules in the outer membranes of gram-negative bacteria, since it may be possible to create novel lipid A structures that are E. coli/Rhizobium hybrids. The specific aims will be; 1) identification of monosaccharide and disaccharide lipid A precursors in living, cells of R. leguminosarum; 2) identification of the enzymes that generate the unusual structural species of lipid A found in R. leguminosarum; 3) isolation of selected Rhizobium mutants defective in lipid A synthesis; and 4) characterization of substructures of lipid A from strains of Rhizobium as endotoxin agonists and/or antagonists.